About Me

I am a final-year Ph.D. candidate in Autonomous Systems & Connectivity at the University of Glasgow, supervised by Dr. Jianglin Lan.

📬 Open for Postdoctoral Opportunities:
I will be available for postdoctoral positions starting Mid 2026, with a focus on structure-aware learning, geometric reinforcement learning, and dynamical systems–inspired decision-making.

💡 My Research Journey

My research began with a fundamental paradox in autonomous driving:

“Why doesn’t better perception necessarily lead to better decisions?”

While working on SLAM, I realized that high-fidelity representation learning alone does not guarantee effective decision-making. This realization pivoted my focus toward motion planning and multi-agent coordination, where I observed that traditional hard-constraint methods often struggle to generalize across the stochasticity of unstructured, open-world scenarios.

This motivated my transition to data-driven approaches, which exposed a deeper, structural limitation in modern Reinforcement Learning (RL): the mismatch between algorithmic assumptions and physical reality. Standard Gaussian policies assume unbounded support, whereas real-world actuators are inherently bounded. This discrepancy leads to instability, inefficient exploration, and suboptimal convergence in high-dimensional continuous control.

To resolve this, I developed a framework that respects the intrinsic geometry of action spaces. Instead of relying on heuristic regularization.

Current Frontier: I am now exploring the theoretical foundations of structure-aware learning. By drawing connections between Reinforcement Learning, Hamiltonian Mechanics, and Symplectic Geometry, I aim to develop policies that are not only performant but also physically consistent, geometrically principled, and intrinsically stable.

I believe the next breakthrough in generalizable AI will come from respecting structure, rather than ignoring it.

🔬 Research Interests

• 🧠 Structure-Aware Reinforcement Learning
  Stability-aware policy optimization and reliable learning mechanisms for safety-critical control.

• 🤖 Multi-Agent Decision Making
  Game-theoretic coordination, level-k reasoning, and scalable MCTS-based planning.

• 🛡️ Safety-Critical Autonomy
  Contingency-aware planning, risk-aware trajectory optimization, and formal safety considerations.

• 🚗 Autonomous Driving Systems
  Interactive decision making in mixed traffic and complex urban environments.

Quick Links

• 📄 View my publications - Research papers and academic contributions
• 🛠️ Explore my projects - Technical implementations and research prototypes
• 🎓 View my CV - Professional experience and qualifications

Recent Updates

• First Author, “Scalable and Safe Multi-Agent Coordination with Reconstructed Level-k Monte Carlo Tree Search” has been accepted to the International Conference on Autonomous Agents and Multi-Agent Systems (AAMAS 2026) (CORE A*). (December 2025)
• First Author, “Uncertainty-Aware Roundabout Navigation: A Switched Decision Framework Integrating Stackelberg Games and Dynamic Potential Fields” has been accepted by IEEE Transactions on Vehicular Technology. (November 2025)
• Co-first Author, “Evaluating scenario-based decision-making for interactive autonomous driving using rational criteria: A survey” has been accepted by IEEE Transactions on Intelligent Transportation Systems. (November 2025)
• First Author, “Safety-Critical Multi-Agent MCTS for Mixed Traffic Coordination at Unsignalized Intersections” has been accepted by IEEE Transactions on Intelligent Transportation Systems. (August 2025)
• First Author, “Contingency-Aware Spatiotemporal Optimization for Safe Autonomous Vehicle Trajectory Planning” has been accepted by IEEE Transactions on Intelligent Transportation Systems. (August 2025)
• First Author, “KAN-LSTM Enhanced Multi-Agent Advantage Actor-Critic Reinforcement Learning for Autonomous Ramp Merging” has been accepted by IEEE Transactions on Vehicular Technology. (July 2025)
• First Author, “A Conflicts-Free, Speed-Lossless KAN-Based Reinforcement Learning Decision System for Interactive Driving in Roundabouts” has been accepted by IEEE Transactions on Intelligent Transportation Systems. (June 2025)
• 🏆 Research Mobility Fund: Awarded £1800 from the University of Glasgow College of Science and Engineering mobility fund for international research collaboration. (April 2025)
• First Author, SLAM2: Simultaneous Localization and Multimode Mapping for indoor dynamic environments has been accepted by Pattern Recognition. (February 2025)
• 🏆 Editor’s Choice Award. My article “Enhanced Visual SLAM for Collision-Free Driving with Lightweight Autonomous Cars” has been selected as an Editor’s Choice Article in Sensors journal. (April 2024)
• First Author, DPL-SLAM: enhancing dynamic point-line SLAM through dense semantic methods has been accepted by IEEE Sensors Journal. (March 2024)

Other Selected Publications (Non-First Author)

• Third Author, “Balanced exploration and attention-inspired decision making for autonomous driving,” has been accepted by IEEE Trans. on Vehicular Technology. (July 2025)
• Third Author, “Balanced reward-inspired reinforcement learning for autonomous vehicle racing,” has been accepted to the Learning for Dynamics and Control (L4DC 2025). (November 2024)
• Third Author, “Efficient and balanced exploration-driven decision making for autonomous racing using local information,” has been accepted by IEEE Trans. on Intelligent Vehicles. (July 2024)